You are currently browsing the category archive for the ‘nuclear power’ category.

I have been reading the database (and associated literature) of the recent IPCC 1.5 degrees special report. I am disappointed that there clearly is no consistent reporting requirements for the IAM models and for example most scenarios do not report critical input assumptions such as capital costs. However, I had a more careful look on how modellers choose to treat uranium supplies. I have noted earlier how some models restrict nuclear power in the scenarios by claiming world runs out of uranium. These seemed very silly to me already then, but now, having read more, they seem totally absurd.

Below are some examples of the used supply curves. Notice how extracting more pushes up the cost of fossil fuels, but how in the case of uranium, the opposite is typically expected. Small increase in price increases the resource base dramatically. (This is harder to see, since axis are switched, but there it is…). Strangely many IAM modellers seem to treat uranium differently from other options. They force a hard upper limit on how much uranium can be extracted. For the German REMIND model this is set at 23Mt of uranium. In others, it might vary depending in the narrative and might be, for example, 20 Mt for the SSP1 (Shared socioeconomic pathway) variants and somewhat higher with other narratives. This resource limit is only applied for nuclear power and there are no material limitations on renewable energy for example. How come?


Uranium supply curve from GEA ,used in several models if I understood correctly. Based on Schneider and Sailor, but with arbitrary cut-offs which vary depending on the desired narrative.

There are some “real” sources that are used to estimate uranium supply curves. There is the IAEA Red book, Bunn et al. paper on the economics of reprocessing vs. once through cycle, and Schneider and Sailor paper on the supply curve. Remarkably, none of these gives support to the claims of uranium scarcity. Modellers have to some extent used the supply curve from the papers, but then added arbitrary ceilings for the supply. Bunn et al. conclude that scarcity will not force transition into breeders since resource base will rapidly expand whenever needed. They also discuss reasons for this extensively. Schneider and Sailor also warn that their supply curves should not be used over generational timescales (as done by the IAM modellers) since they ignore learning effects. They anticipate, based on historical data, that if mining volumes increase, prizes will drop rather than rise. This is precisely opposite to what is assumed in the scarcity based IAM models. IAM modellers simply ignore all this.

So are the assumed uranium limits relevant or not? Over short time scales, they are obviously not, but over the century they are. To illustrate this, I checked from all the scenarios in the 1.5 database how much uranium they would (roughly) need and compared this to the 23Mt limit assumed in the REMIND model. There are some models (GCAM in particular) where massive amounts of nuclear power are constructed and the uranium limit doesn’t seem to matter a lot (see the peak in the figure below). However, for most of the scenarios the required amount of uranium is suspiciously close to the upper boundaries set as input assumptions.


Estimate of how much uranium relative to the 23Mt limit assumed in the REMIND model IPCC scenarios need.  Some models with only few scenarios are not marked. Flag indicates “continent”. I did not feel like splitting the European models further.

What does this mean in terms of the energy consumption of humanity during the next century? What have the modellers actually assumed? So let me add rough estimate of the assumed uranium supply curve into the supply curves for fossil fuels. This requires some unit conversions and I also convert from nuclear electricity to primary energy by dividing it by 0.33 to get a comparable cost per GJ. As you can see in the figure, the underlying assumption, for example, REMIND modellers use as an input, is that uranium is the first resource to “run out”. Constraint is so strict that nuclear power can only cover a small fraction of the energy consumption (something around 50-100ZJ maybe?) over the century.

For some weird reason, humanity stops mining uranium even when the fuel cost is still massively lower than for fossil fuels. What is going on? Why would this make any sense? The used references certainly do not support this. (Incidentally, is this why uranium supply curve is given in different units …to make this comparison harder? I hope not.) Many people read these scenarios as outcomes of “science”. Computer magically optimizes something and then gives us a guideline on how to behave and what decisions to take. This is not at all the case. Scenario modelling has its uses, but mainly in illustrating sensitivities etc. If silly constraints are imposed from the outset, silly outcomes will appear. Garbage in, garbage out.


The lowest blue line is my estimate for uranium in light water reactors. It ends when roughly  23Mt of U are used. (I hope I made the estimate right. I encourage you to double check.) Wow!

Added 3.11.2018: I did a similar check on the database for the 5th assesment report. Figure below. (I wonder why REMIND clustered around 1 earlier, but now more like 0.5? What changed? The REMIND scenarios clustering around 1/2 seem to change their data reporting from every 5 years to every decade at 2060. Could that have something to do with this? Curious.)


Same as earlier, but for the AR5. Not quite all the models/scenarios are included, but most are. I also added a small indicator to distinguish Message v.4 runs from the rest since it seemed behavior was rather different.

A Greenpeace report commissioned from 100%RE academics from Lappeenranta University of Technology (of course) on electricity generation costs was recently published in Journal of Cleaner Production. Details of the computations were kindly made available as a supplementary spreadsheet. The results left something to be desired. I wrote a response which has now been published. “Response to ‘A comparative analysis of electricity generation costs from renewable, fossil fuel and nuclear sources in G20 countries for the period 2015–2030’” (doi:10.1016/j.jclepro.2018.07.159). Link should work for few months after which you either need access to the journal or download it from here. Take away lesson as usual: Be critical and do not automatically trust the results or conclusions.

Added 18.10.2018: The free share link does not seem to work from WordPress so if you don’t have access otherwise you can download it from here.

I ran into “Roadmaps towards sustainable energy futures”-project. It is a german funded project which has created a set of different scenarios for IAM simulations. They describe themselves:

“The Roadmaps towards Sustainable Energy futures (RoSE) project aims to provide a robust picture on energy sector transformation scenarios for reaching ambitious climate targets. A broad and systematic exploration of decarbonization scenarios for the energy system is indispensable for better understanding the prospects of achieving long-tern climate protection targets. RoSE is assessing the feasibility and costs of climate mitigation goals across different models, different policy regimes, and different reference assumptions relating to future population growth, economic development and fossil fuel availability, in order to provide vital insights into the overarching policy question: What are robust roadmaps for achieving a sustainable global energy future?”

Now I am not really a big fun of these modelling games since one tends to get out whatever one puts in and I have had a feeling that not all modellers model carefully enough. See my earlier post on this… However, ROSE-project has a database for the modelling results and let me just show what the 450ppm medium growth, medium population growth, and medium convergence scenarios ROSE211 give for the primary energy supply. I show bioenergy, non-bio renewables, and nuclear.


Insane amounts of bioenergy in all except IPAC


non-bio RE grows a lot. REMIND model extreme is this.


Wow! Colossal increase in nuclear generation from todays abut 10EJ/year. In REMIND reductions in the 2nd half of the century…for silly reasons.

See how two of the models actually see about 20-fold increase in nuclear generation. In those model scenarios capacity growth in the 2nd half of the century seems to be more than 100 GW/year (probably unrealistic). If I understood correctly REMIND model is a german model and even that sees dramatic increase in nuclear power until about 2060 after which it declines. They got this result by forcing it down by declaring world runs out of uranium at this time. This is an assumption other presumably considered silly, but which does have the benefit of creating a safe-space for german IAM modellers. I find it curious to observe the disconnect between the modelling results and how they are communicated in public.

I have written before how IEA sometimes hides politically inconvenient results in their reports. Now IRENA has published a new report “REmap: Roadmap for A Renewable Energy Future: 2016 Edition”. It naturally has a myopic focus on renewables which is the purpose of the organization. Nevertheless I was somewhat interested in their cost and capacity figures. In the next 15 years their REmap plan calls for a total investment of about 6000$ billion into wind and solar.

What does this spending buy? According to IRENA it buys about 1600 GW of wind power capacity, 1585 GW of PV capacity, and an explosion of installations into concentrated solar power so that its capacity would increase from about 4GW today to 110 GW at 2030.


If we assume an average global capacity factor for wind power to be about 25%, for PV about 15%, and CSP about 40%, we find that added wind and solar production corresponds to about 690 GWe of continuous production over the year. With 25 year lifetime, this means an electricity production of about 151 PWh. So without discounting etc. we would pay about 8.7$ billion/GWe for delivered (average) power. Capital costs would imply (without discounting and other expenses) about 4 cents/kWh cost of electricity.

IPCC 5th assessment report median nuclear overnight costs 4300$/kW. Let me again ask the naughty question since IRENA refused to compare options (1,5). What could we get, if we were to plough the money IRENA desires to spend for wind and solar into nuclear? I will round the cost to 5000$/kW for nicer figures. It really doesn’t matter. We could buy 1200GW of capacity which implies about 1100GW production at 90% capacity factor. Much more than than the 690GW IRENA bought. With 60 year lifetime, the actual production and reduced emissions are larger by a factor of about 4 and correspondingly the cost per kWh from non-discounted capital costs is around 1 cent. Estimates are so far apart that fiddling with details is not going to change anything.

Savings from external costs according to IRENA

Savings from external costs according to IRENA

IRENA also finds that their plan costs more than reference scenario (which is also not a cost minimizing option), but makes it alright by assigning externalities to the reference case (10,11, and 12). Outdoor and indoor pollution would be reduced by poor people burning less dung and biomass and some (smaller) savings also appear from reduced CO2 emissions. These are all savings that can just as well be assigned to the nuclear build-up sketched here except that savings would be considerably higher by hundreds of billions every year. Just sayin… (Of course I understand that at this point rules must somehow be changed.)

IRENA also states:

“Avoided investments in non-renewable power capacity alone are estimated at USD 1.5 trillion to 2030, or about USD 100 billion per year on average in the 15-year time period. Almost half of these savings would come from not building coal-fired power plants; another 30% from nuclear investments seen as no longer necessary. “

Mind boggles. No longer seen as necessary since authors proved themselves willing to impose additional costs on others? You do see that in the plan I outlined we would get much more savings in avoided investments than in the IRENA plan? Why settle for lower emission reductions? Have we been reducing emissions too rapidly? If you want to promote wind and solar, that is fine with me. But could you please make a case that somehow makes sense? Claiming that plans are economical even when it is manifestly clear they are anything but, undermines your message outside your echo chamber. Hopefully the plan is not dependent on everybody living in the same chamber. Not really my cup of tea. I rather stand in the rain outside.

P.S. Justifying climate action with external costs from indoor biomass burning and outdoor pollution is a dubious idea. Most of those costs can also be avoided by switching from dung and biomass to fossil fuels especially if appropriate pollution controls are used. Implicitly IRENA et al. base their logic on things NOT improving outside their chosen set of tools. This makes no sense.

I glanced at the IEA report “Energy technology perspectives 2012”. (There is also a 2015 version, but I didn’t have access to that. Annoyingly IEA charges dearly for these reports so that even though they are commonly referred to in discussions, they are not widely available.) In their baseline 2DS scenario IEA estimates cumulative saving (savings in fuel minus investment costs) over 6DS scenario of 26 trillion US dollars by 2050. Interestingly enough they also have a “high” nuclear scenario where they tolerate more nuclear power than in the baseline. In this scenario the savings are largest, 27.9 trillion. Strangely enough this result was buried to the page 384 of the report. Wouldn’t it have been useful to highlight this since there are plenty of people and (believe it or not) politicians who don’t know this? After all this ignorance might make them promote policies that are counter productive in fighting climate change.

IEA_ETP2012_TableOfScen 2015-06-02 12:42:18_mod

We can also look at the required investment level to follow the 2DS scenario. Here IEA assumes large cost reductions for renewable energy sources. This might or might not happen, but let us just accept this for now. I highlight some relevant numbers from the report in the following table.

IEA_ETP2012_NeededInvestments2015-06-02 12:34:38

Source Investment 2030-2050 per year (billion) Production 2050 (TWh) TWh/billion
Nuclear 119 7918 66.5
Wind 167 6145 36.8
Solar 232 5988 25.8
Wind+solar 399 12133 30.4


As you can see, even though IEA has baked in massive cost reductions assumptions into solar and wind by 2050, they still deliver less than half as much electricity per investment than nuclear (for which IEA didn’t seem to assume learning effects). IEA 2DS baseline is not a cost minimizing scenario, but presumably reflects sufficiently conventional wisdom that authors believe is more palatable for IEA funders.

What would happen if we were to simply divert investments spent from more costly decarbonization options to nuclear? That 399 billion for wind and solar would then enable about 14000 TWh/year more carbon free electricity than the 2DS baseline. This would be enough to eliminate coal, coal+CCS, natural gas, natural gas+CCS, biomass+waste, and biomass+CCS from the electricity mix at 2050 with more than 1000 TWh left over. As these sources of electricity disappear, more than 100 billion a year is also saved in investment costs and lots more in fuel costs. Based on the difference between IEA 4DS and 2DS scenarios, I estimate around 20 trillion additional cumulative savings in fuel costs. Not bad, I would say given the speculative nature of CCS technology, environmental and social impacts of bioenergy schemes, and the need to decarbonize also other sectors than electricity production. (Incidentally, it tells something of the absurdity of current energy discussions, that many celebrate large investments as a good thing. It doesn’t seem to matter what the investment actually buys. More expensive the better, because that means more investment and larger business opportunities in “cleantech”.)

Do I think this will happen in the near future? Of course I don’t. If there would be a wartime-like urgency, who knows, but as it stands such scale up is not going to happen. However, even if unrealistic this option is MORE realistic than the renewables-only party line. It is more realistic economically, technically, and in terms of material limitations. Since it is not going to happen, (as I have said many times before) we can look forward to much more than 2 degrees warming.

In its latest assessment report IPCC concluded that in order to get climate change under control world needs massive expansion of nuclear power, renewables, energy efficiency, and CCS. I am a numbers guy and therefore I was delighted when I found a useful database for many of the mitigation scenarios IPCC relied on in its latest report. There is a database for the scenarios and additional information and assumptions used on many scenarios can be found in another database. I found this very interesting since articles reporting on the scenarios often explain the underlying assumptions of the models poorly. I will focus now on how the modellers approached nuclear power. I didn’t have the patience to go through all scenarios and I focused on those with 450ppm CO2 target that contained all technologies optimally (allegedly). I found that quite a few modellers dealt with nuclear power in a way that left me wondering if their modelling is simply poorly disguised ideological propaganda.

Some main approaches used to influence how well nuclear power does in the models relative to variable renewables (wind and solar):

  1. In many models nuclear capacity increases massively. Hundreds and hundreds of reactors are constructed, but amazingly nobody learns anything! Capital costs for nuclear power are typically kept almost constant throughout the decarbonization pathways. On the other hand learning effects and technological evolution are assumed for other energy sources. For wind and solar power these are often assumed to be very dramatic and there are learning effects even for fossil fuels. So this tough love only seems to apply to nuclear power.
  2. Many models assume large cost reductions for wind and solar. In the end, this is not much more than a wishful guess.
  3. Some models assume anomalously large capacity factors for wind and solar. See for example, “Message Ampere2-450-FullTech-OPT” scenario. Capacity factors for wind are almost 40% while for solar power they use about 25-31% over the course of the century. Since real figures are more like half of the assumed figures, the model drastically underestimates the costs for wind and solar. (IMACLIM scenarios seem to do the same)
  4.  Some models (IMACLIM in particular) assume very low capacity factor for nuclear.  “IMACLIM Ampere2-450-FullTech-OPT” has a nuclear capacity factor of just 45% in 2100 while for wind and solar they have 36% and 38% respectively! This doesn’t just roughly double the cost of nuclear in these models, but also underestimates the costs for wind and solar.
  5. Some models (REMIND and MERGE-ETL) postulate a world running out of uranium together with no technology development for nuclear. This “peak uranium” then limits the role nuclear power plays in decarbonization.

Figure 1: Nuclear power in Remind Ampere2-450-FullTech-OPT scenario. Massive increase and then…

Let me discuss the sillyness of the last trick in more detail. Figure 1 shows what REMIND scenario got for nuclear power when all technologies were used “optimally”.  So massive increase in nuclear power until middle of the century and then rapid decline. Decline is caused by uranium supplies running out as soon as light water reactors with once-through fuel cycle have used 23 million tons of uranium. This is very strange for several reasons.

First, this number doesn’t seem to bear any clear connection to known uranium resources which are about third of this figure. Modellers probably felt that using known resources as an upper limit would have been too stupid to pass the laugh test.

Second, mineral resources have a habit of increasing together with demand since increasing demand stimulates increasing investment in exploration and technology development.  In the past one hundred years copper production has increased by an order of magnitude. All this time world has been “running out” of copper in about 40 years. Uranium is not especially rare element and there is no reason to believe we are running out of it anymore than we have for other metals such as tin which has about the same crustal abundance.

Third, from where does the assumption of no technology development come from? Wasn’t this supposed to be a scenario where all technologies are allowed? For nuclear power technologies that that improve the fuel efficiency by about two orders of magnitude are already known.

Fourth, why is there resource constraint only for nuclear power? The resource constraints are more severe for wind and solar power (and for bioenergy). In Figure 2 I show an image I picked up from a european study on critical metals for energy technologies. The elements with greatest supply risks are used in the construction of wind and solar power. (By the way, the only nuclear related element on the list is the low risk hafnium for control rods.) Figure 3 I picked up from a fairly recent Alonso et al. paper. Authors estimated that dysprosium (used in magnets) demand in renewables heavy mitigation scenarios is expected to be a whopping 2600% higher than projected supply already in 2035!


Figure 2: Critical metals for European “strategic energy technologies” according to European commission Joint research centre study.

Figure 3: Expected demand and supply for Dysprosium according to Alonso et al.

Figure 3: Expected demand and supply for dysprosium according to Alonso et al. (2012).

What would happen if we were to apply modellers approach for renewables? Let us just take silver as an example. Silver reserves are estimated at about 530000 tons. Let us assume that “real” resource is 4 times this (remember uranium resource was set at 3 times the known reserves) and that half of this can be used for photovoltaics. There are after all other uses for silver as well. Since 1GW of solar power requires about 80 tons of silver, this means that at maximum we can have about 13TW of solar capacity as opposed to almost 90TW cumulative capacity REMIND modellers extrapolated. Instead of being the largest contributor to the primary energy supply its contribution would fall into 5-10% range. The amount of silver required to construct the solar power in REMIND FullTech scenario is about 13 times larger than the estimated global silver reserves. Now can there be ways around these constraints? Probably there are and maybe we could use less silver, but using substitutes might imply higher costs and worse performance and furthermore, if one was not permitted to use already demonstrated technologies for nuclear power why should imaginary advances be permitted for other alternatives?

What might we get if we remove this silly constraint from the model? Obviously I cannot repeat the exercise with the tools I have available, but we can get a rough estimate. Lets take the growth rate (4.8%) for nuclear power REMIND modellers established between 2020-2050 and just let it grow with the same rate until the end of the century. This is not extraordinary in the context of this model since for wind+solar the growth rate through the century was 7.6% even though capital costs are such the nuclear power seems to have a lower levelized cost of energy (5% discount) throughout the decarbonization pathway. I show the result in Figure 4. Nuclear power would end up dominating the energy supply.

I have a feeling that resource constraint was introduced specifically for this reason. Modellers first did their calculations without the constraint and ended up with a result that they found distasteful. They did not want to go on record with the scenario that might “rock the boat” or give people funny ideas. By introducing the resource limitation for nuclear power they could clip its wings and keep it supposedly as an option while limiting its role to the margin. In fact that strange 23 mton uranium resource limit seems to suggest that over the century LWR:s cannot produce more than maybe around 5% of the primary energy. I suspect that modellers worked backwards and set the resource limitation based on the maximum share of the energy supply they were ready to grant for nuclear power. Not cool.

Figure 4: There, I fixed it!

Figure 4: There, I fixed it!

Then there is PRIMES…sigh. This is a model I encountered few years ago as I was reading EU:s 2050 energy strategy. I remember glancing at the referee report and being troubled by the brief remark on page 6. Referee had asked about rather optimistic cost assumptions to which response was that if capital costs for wind are set higher then the future learning curve can be steeper. To me this suggested that modellers were perhaps fitting model to the fantasy. In the AMPERE database PRIMES scenarios for EU are also included. I was naturally most interested in the Ampere5-Decarb-AllOptions scenario which according to authors is a scenario “with all technological decarbonisation options available and used according to cost optimality; this scenario provides the least cost decarbonisation pathway for the EU.” Sounds interesting! However, as you look at the actual results you notice something weird. The capital costs assumed are such that nuclear (again) has the lowest LCOE throughout the decarbonization pathway. Despite this modellers claim that nuclear generation in EU will decline by 20% by 2050. How is this even possible?

Then I noticed a strange footnote on page 15: “PRIMES assumes that nuclear development has been significantly affected in the aftermath of the nuclear accident in Fukushima in March 2011. Both PRIMES and TIMES-PanEu impose national constraints regarding nuclear, such as countries’ decisions not to use nuclear power at all…” Please tell me that I am reading this wrong. They didn’t just exclude nuclear power from large parts of EU in their “all options” scenario for political reasons and then sell it as the cost optimal one?

I have now outlined several ways in which scenario modellers seem to suppress nuclear power from their reference scenarios where all options and technologies are supposedly on the table. This has also consequences for the other scenarios and comparisons between them. Since modellers suppressed nuclear power already in “the tech neutral” scenarios adding additional anti-nuclear policy, can be presented as not really having major cost consequences.

Figure 2: The box on the left has nuclear power in it and the box on the right had it removed. Amazingly it looks almost the same as the other empty box!

Figure 2: The empty box on the left has nuclear power in it and the box on the right had it removed. Amazingly it looks almost the same as the other empty box!

Since I am a bad boy I will conclude with some rough estimates on what would it take to replace (gasp!) solar and wind power at the end of the model scenarios with nuclear power that generates the same amount of electricity. I simply estimate the required nuclear capacity (90% CF) and use modellers assumptions about capital costs. Required yearly outlay is roughly total capital required divided by the lifetime of the plant. I will use 30 year lifetime for wind and solar and 60 years for nuclear. (Numbers are in billions of 2005$…I think.)

Model Wind+solar capital Nuclear capital (Wind+solar)/year Nuclear/year
Remind 450-FullTech-OPT 74540 62753 2485 1046
Message 450-FullTech-OPT 40620 64150 1354 1070
IMACLIM 450-FullTech-OPT 5680 5765 189 96
Primes Decarb-AllOptions (EU) 1430 826 48 14
Primes HIEFF-NoCCS-NoNUKE (EU) 1555 900 52 15

In all models the required yearly outlay (at 2100 or 2050 for PRIMES) for energy supply is dramatically lower if we replace wind and solar capacity with nuclear power. This despite the fact that MESSAGE and IMACLIM assumed unrealistically high capacity factors for variable renewables. It is remarkable than even though this kind of chicanery was going on behind many models, IPCC still ended up concluding that nuclear power must expand massively. This is perhaps partly because not all scenario builders were intellectually dishonest about this issue and some models ended up, for example, with ten fold increases in nuclear capacity. On the other hand I am afraid that all 450ppm scenarios are utterly unrealistic….and don’t get me started on their absurd bioenergy projections.

P.S. I spent some time copying the data I was interested in from the database. Interface seems a bit uncomfortable for that. Here is a link to some of the data I extracted.

P.P.S.  For laughs you might want to check IMACLIM model with 550 ppm goal and CCS excluded. Since the original one was very strongly dependent on CCS one would imagine that ruling it out would have interesting consequences for the energy mix. See what modelers assumed for the capital costs of nuclear here to suppress that out of control (critical?) nuclear growth early in the century.



Note added 6.10.2019: It would seem that the capital cost data for the AR5 scenarios has been removed from the database. At least I cannot find it anymore. Can someone explain where is it now?

There was a brief, but interesting discussion in Twitter about risks from exposure to low levels of ionizing radiation. Among pro-nuclear people this discussion erupts with some regularity. For some background there is this really clear discussion by @kasilas which you should read. The thing is that some (I suspect mostly people with engineering background) dislike LNT (linear no threshold) assumption in radiation protection. They say that below a dose of about 100 mSv it doesn’t have observational support and therefore one should not talk about “risk” below some threshold. Such risk is speculative and just gives ammo to anti-nuclear crackpots. On the other hand experts in radiation biology and protection gather around the “party line” and tend to see LNT, if not perfect, then at least good enough and certainly better justified than supposed alternatives. The sane on both sides nevertheless conclude that whatever risk model we use for low doses, the risks will  be small compared to many other risks we face on a routine basis. Both, by and large, hold the opinion that radiation from nuclear power is not an important public health concern relative to more pressing concerns.

Figure 1: Discussing hormesis and how it relates to LNT

Figure 1: Discussing hormesis and how it relates to LNT

I think this discussion is interesting not so much from the scientific perspective, but mainly from the sociological perspective. I suspect that engineering types dislike going through the trouble of minimizing all sources of exposure as much as possible while knowing that it adds to costs and that this work has no observable consequences. They feel that they could be working on much more important things. Radiation protection people on the other wish to protect scientific standards and probably feel a civic duty to maintain and built public trust on experts. Playing fast and loose with radiation risks might undermine that work. They dislike fear mongering by anti-nuclear folks as well as nonchalant attitude to small doses expressed by some pro-nuclear people. They are the doctors trying to keep inmates from running the asylum. (Although this task is complicated by the fact that only pro-nuclear folks have the courtesy to loiter close to the asylum. Antis have always been running free.)

Personally I have sympathy for both sides of this discussion, but I think this is fundamentally not a scientific question, but a question of public perception of risks and how that relates to policies. Due to decades of misinformation many people have fundamentally wrong perception of radiation risks. When we start by saying that radiation dose, no matter how small, poses a risk, we do not question that underlying default setting. We might then continue telling how this risk is nevertheless tiny, but many people have already tuned out. And in any case people are very bad at evaluating risks so they are more than likely to compress the message to “radiation BAD”. The conspiracy minded among the public will of course go even further. When official tells them small amount of radiation has risks, they will conclude that it is in fact deadly and the level that is really safe will be something much much lower. As the safety level is thus adjusted downwards possibilities for exceeding those “safe levels” multiply and the sense of danger will probably go up rather than down. Of course this is a complex issue. If on the other hand we say that the risk is not there, some will simply decide that you are not credible and tune out immediately. You have to adjust your message in response to craziness on the other side and hope they will gradually move to a sensible position. But does anybody know, how nuanced accurate discussion actually influences people whose opinions are at the start of the discussion bizarrely off base? Such discussion certainly is preferable with people whose opinions are more or less sensible to begin with, but with others? I am really not sure and would love to learn of some research on this topic.

Given my background I was (of course) thinking that isn’t this kind of similar to importance of quantum mechanics? We live in an imperfect world where most people do not need Planck’s constant in their daily lives. This natural constant is at the heart of quantum mechanics and indeed our world be inexplicable without it. (In fact some of those who actually need it in their daily lives, define their units in terms of it so that for them Planck’s constant has a value one. Being so down to earth and organic they even call such units “natural”.) However, as a practical matter it doesn’t make sense to incorporate the effects of Planck’s constant into building codes or environmental impact assessments etc. Most people will find it easier to just set Planck’s constant to zero and as a practical tool that is usually perfectly OK, even though it is fundamentally wrong. In fact, if we were to do the opposite, the risk of a backfire would be large. People would not know how to deal with Planck’s constant in practice and if asked about its magnitude they would be off by a large amount. (If we were to give them some additional information such that “Planck’s constant is related to the energy of  particles of radiation”, many would probably increase the value of the constant even more.)

Given the horrendously wrong public perception of radiation risks, I often feel they would be better served if their default settings were based on the idea of zero risk. This is fundamentally wrong, but it is less wrong, in a practical sense, than their current perceptions. Once the lowest order term has been correctly established we could start adding nuance and even move to discussion of such regimes where radiation risk is actually large. Nowadays people start from fears of cities attacked with nuclear weapons and then we expect them to make a reasonable extrapolation of risks into their daily lives. For most people I don’t think that will ever happen. On the other hand, I do not know how that more sensible starting point can be established in practice. Currently people pickup nonsense from NGO:s and media already as children and accurate information gets drowned in the noise.

Recently a report on energy costs prepared for EU commission by the consulting company Ecofys crossed the news threshold in many places. Usually it has been reported as being “the EU report”, but EU commision states “The views have not been adopted or in any way approved by the European Commission and should not be relied upon as a statement of the European Commission’s views. The European Commission does not guarantee the accuracy of the information given in the studies, nor does it accept responsibility for any use made thereof.” So the report has not been “endorsed” by EU commision (although paying Ecofys for the report is bad enough). Ecofys did the work on WWF bioenergy-heavy renewables-only energy vision and is widely linked and quoted by environmentalists in Europe.

Following quote from WWF report captures quite well, why I am not a fan. “Ecofys estimates that we would need around  250 million hectares of agriculture land,  which is equivalent to about one-sixth of  the total global cropland today, as well  as 4.5 billion cubic metres of biomass from already disturbed forests. But what  is possible on paper, even after the most  rigorous analysis, is a different matter in  practice. We have yet to identify where  this land is, and how it is being used at the moment.“: WWF. This is then followed by WWF nevertheless endorsing such a vision.  I had a look at this new report. Below few comments.

  1. As you can see from the Figure 1, they find that nuclear power is not heavily subsidised and is among energy sources with low external costs. According to Ecofys external costs are only little higher than with the worst renewable (biomass). This is not news, but it is interesting that even Ecofys is forced to acknowledge this. (See later for their desperate attempts to change the results…)

    Figure 1: Summary of costs, subsidies, and external costs

    Figure 1: Summary of costs, subsidies, and external costs

  2. It has been widely quoted (example here) that according to this report wind power is cheapest source of energy. It is perhaps helpful to note that  Ecofys gets this result by discounting nuclear costs with 9-11% rate while discounting wind power with a much lower rate of 5-7%.
  3. As you can see from Figure 2, according to Ecofys external costs of nuclear is dominated by “depletion of energy resources” category. This was very strange result.
    Figure 2: External costs

    Figure 2: External costs

    It turns out that they calculate a cost of depletion as 0.05 euros/(kg oil. eq.) both for fossil fuels and nuclear power! Ecofys used a tool called “Recipe” to calculate external costs and interestingly in case of nuclear power they decided to specifically deviate from what developers of Recipe said was the appropriate methodology. ” Unlike metals, we cannot use the concept of grade to express the quality of oil and gas resources. Conventional oil and gas will simply flow out of the well up to a certain point. After that point is reached it is still possible to extract more, but this will increase the production costs and the production energy requirement. Once the energy price increases, it also becomes possible to extract other unconventional resources, such as tar sands, the use of gas liquids, converting gas to oil or coal to oil etc. This means the increase of costs and energy is not caused by a gradual decrease of ore grade, but because more and more mankind will have to switch from conventional resources to unconventional resources…Uranium was formed in the same way as all other metals, the characterisation factor for Uranium is thus included in the impact category for mineral depletion and not fossil fuels.” Recipe in fact gives external cost for Uranium extraction and finds that it is similar to oil per kg.  However, since energy density is different by a factor of 10000-million (roughly…who cares) depending on reactor type, Ecofys inflated otherwise irrelevant externality into one that dominates external costs of nuclear power. Not cool. Perhaps they did this in order to inflate the external cost of nuclear power to be at least higher than renewables they promote?

  4. Historical subsidies for nuclear are mostly based on the idea that state participation lowered interest rates for the projects and that difference between imagined market rate and state interest rate constitutes a subsidy. The logic here is not convincing and also requires the value choice that “market interest rate” is the correct one and states participation is an interference into natural order of things.  Whatever your opinion happens to be on that one it is important no notice, that when Ecofys calculates external costs for depletion of resources they assume owners of resources use too high discount rate and that socially optimal one is lower. So now the market no longer knows better. On the other hand when they calculate the levelized cost of energy (LCOE) they use different interest rates for different technologies.  Maybe it is true that some wind power developer can get cheaper loan from the bank, but they do so because state has guaranteed them customers as well as the price with feed in tariffs.   According to Ecofys such political interference was supposed to be a subsidy and real interest rate was the one without political support structures. In case of nuclear power it is the political uncertainty that increases the perceived risks and consequently it is suffering a “negative subsidy” due to politics. Strangely here Ecofys nevertheless takes interest rates as “correct ones” rather than interpreting the resulting LCOE as the one after political interference. So note how the ground keeps shifting, but always in such a way as to inflate costs for nuclear power and fossil fuels.
  5. ECOFYS ignores some external costs. For example, there is an external cost for occupying agricultural land (0.1 €/m^2). This cost is due to monetizing the lower biodiversity of agricultural land as opposed to natural habitat.

    Figure 3: External costs? What external costs?

    However, there is no cost that I can see associated with removing biomass from forests for burning. They rationalize this by “We assumed wood pellets are made of residue wood and did not allocate agricultural land  occupation to the production of this wood …”  So a precondition for this resource is a forest industry creating huge externality and “waste” stream, but none of this is reflected as an externality for bioenergy.

  6. Existence of higher system level costs from intermittent renewables is acknowledged in the text, but these are not counted as costs, subsidies, nor as external costs. They are, as far as I can see, simply not included in any category.
  7. After renewables subsidies (41 billion euros/year) largest subsidy category (27 billion) was for “Energy demand support”. This is almost entirely due to lower tax rate for some uses of fossil fuels. I think this is the way also OECD defines subsidies, but in my opinion it is deeply misleading. If I am not taxed according to maximum rate, am I receiving a subsidy? If we use the Ecofys definition for energy subsidies, yes I am. Where I live (Finland) state gets more than 4 billion euros income from energy taxes that mostly tax fossil fuel use. However, this is not counted as a “negative subsidy” for fossil fuels. If we would stop burning oil, state would lose billions in tax revenue. If oil burning is then replaced with some other energy source requiring subsidies (for example) of 5 cents/kWh, state would need to find billions more. In total such transition could easily cost the state as much as we spend on education, but when computing subsidies there would have been no change. Definition is insane.

IMG_1276.JPG Sooner or later we will have to decarbonize also transport and it is quite likely that it will involve production of synthetic fuels using carbon free electricity as an input. Let us think a bit what this implies for the power source used to power this production. (For some related thoughts with respect to electricity storage see here.)

Imagine an option A, where we built so many nuclear power plants that they meet maximum demand (electricity sector has then been decarbonized) and then use excess power to run plants producing synthetic fuels. In option B we will do the same, but with wind power. We will built so many turbines that they produce the same amount of electricity as nuclear power plants in option A. We will use the excess again to produce synthetic fuels. Since I have the production and demand date for United-Kingdom easily available, I will use that (for the year 2013) to give me characteristic production and demand profiles.


Fig 1: A plant producing synthetic fuels will receive this electrical power as an input. In the lower figure there is also a line indicating the required backup generation for those periods when wind is not adequate to meet the demand.

The following table gives estimates (based on UK figures) for the required capacity for electrical generation, synthetic fuel plant capacity, utilization rate of plant capacity, amount of electricity plants have available over the year (presumably proportional to synthetic fuel production), and required (dispatchable) backup capacity. As is clear, synthetic fuel plants working with wind power have a much lower capacity utilization rate since they have to be able to process much higher electrical powers even though that peak power is rarely available.

Option A Option B
Capacity 63 GW 163 GW
Plant capacity 36 132
Utilization rate 56% 20%
Electricity input 180 TWh 231 TWh
Backup 0 W 47 GW

From these we can also calculate estimates for what the synthethic fuels might cost. Here my only interest is on the RELATIVE cost of options A and B so do not take specific numbers too seriously. For concreteness I assume that synthetic fuel plant will cost 3 billion/GW, has a life time of 40 years, and costs are dicounted with 5% discount rate. (Numbers are made up just for the sake of comparing options…that is all.) I could imagine that plants coupled to wind power might have lower costs per gigawatt due to economies of scale, but since they have to cope with less reliable input power, in first approximation, it feels fair to assume same capital costs for plants. I also assume that operation and maintenance costs are the same so only difference is in the characteristics of the source of the input power.

Fig. 2 shows the result so that on the x-axis we have the relative share of operation and maintenance costs. It is clear that plants coupled to nuclear power plants can produce synthetic fuels at  much lower cost. This difference is mainly caused by the higher capacity utilization rates they enable. If the goal is to displace oil, it will happen easier with an option whose costs are lower. (Not that this goal interests all.)


Fig. 2: Cost comparison for options A and B (Never mind the units on the y-axis…)

Of course we could imagine building power generation capacity that is intended only to run plants producing synthetic fuels. Figure 3 demonstrates this option and shows that it doesn’t change anything of relevance in the comparison.

Fig. 3: Use all power generation for fuel production. Base load vs. variable

Fig. 3: Use all power generation for fuel production. Base load vs. variable

It should be noted that these differences will not disappear with political will or technical progress. Characteristics of the power source has cost implications for the user of electricity (plant, somebody providing services, consumer…). Producers who have access to reliable power supply can outcompete others since higher capacity utilization rates are enabled (and more reliable operations in general). This is an obvious point, but strangely few seem to realize this. Here I used synthetic fuel production as an example, but obviously the argument is equally valid for any production that uses electricity as an input. Even if electricity would be free costs for the consumers are not the same. Cost is not the same thing as value.

I will end with a brief disclaimer. Based on John Morgan’s estimates, in option A UK could produce less that 5% of the oil consumption using excess power. If synthetic fuel production is to play a significant role, electricity production must increase drastically.

Heysham nuclear power plant

In the second week of august EDF decided to shutdown their reactors in Heysham and Hartlepool.This was a precautionary measure after finding a defect in the boiler of Heysham unit 1. In total 4 reactors with total capacity of about 2.6 GW have been taken offline. Some were quick to declare that wind power came to the rescue when nuclear power was proven unreliable (for example Ari Phillips in Thinkprogress, Greenpeace, Giles Parkinson in…)  More recently Justin McKeating from Greenpeace repeated the claim.
…we see a reversal of the view that renewables need to be supported by nuclear power. Although nuclear and wind power do not have the same generation characteristics, nuclear reactors now needing to lean on renewables means the nuclear industry has a big problem.” Given that the claim appears unlikely on meteorological grounds and no evidence for it was provided, I felt a more careful scrutiny was called for.

So, did wind power replace missing nuclear capacity? Short answer is, no it did not. Missing nuclear generation was mostly replaced by increasing use of coal. In Figure 1 I show the output of relevant power sources in UK between Saturday 9.8 and Thursday 14.8. EDF reactors were ramped down during this period and this can be clearly seen in the figure. Equally clear is that when nuclear output was declining, wind power output was declining even more steeply. So rather than coming to the rescue, wind power was unfortunately galloping away when the action started. The reduction in the amount of wind and nuclear power, was mirrored by a clear increase in gas and coal power. Contrary to earlier claims, low carbon sources were replaced by fossil fuels.

Fig 1: UK power production during the reactor shutdows.

Fig 1: UK power production during the reactor shutdowns.

This quick check does provide the answer to our specific question, but with the data available we can learn more. In the following table I show the average power levels for the most relevant quantities shortly before and after the shutdowns. Most pronounced changes were in the amounts of power derived from fossil fuels, nuclear power, and wind power. There has also been some increase in hydropower generation.Recent weeks have in fact been more windy (not unusually so) than weeks prior to shutdowns and power generation from fossil fuels has increased slightly. However, as the earlier figure makes clear, to understand which power source is replacing which one must look deeper than averages.

Period 27.7-9.8 [GW] 14.8-28.8 [GW] Change [GW]
Demand 30.4 30.3 -0.1
Nuclear 7.9 6.0 -1.9
Wind 1.3 2.3 +1.0
Gas 12.4 11.5 -0.9
Coal 4.6 6.1 +1.5
Interconnects 2.6 2.6 0

To get a clearer insight as to how different power sources are connected in UK, we can inspect the data for the year 2013. Figure 2 shows the scatter plot of wind vs. gas for one month period in 2013 together with a least square fitted line. When wind generation is high, gas generation tends to be lower by almost the same amount as wind power generation. The color indicates power demand at that moment. As is clear there is clear gradient towards red with increasing use of gas demonstrating how gas power is used to meet increasing demand. Similar gradient is missing as wind power output increases. (Why did I take one month sample? If we do similar exercise over whole year, we will find spurious correlations between power sources since power demand has seasonal variability so that it peaks in the winter. Wind power production also tends to be higher in the winter. These suggest that if looked over the year, increasing amount of wind power would imply increasing coal and nuclear generation as well. This is clearly nonsense and correlation is caused by increasing base load demand in the winter and scheduled maintenance of nuclear power plants in the summer which happen to correlate with wind speeds. Aggregating the data to monthly sets removes most of these artifacts.)

Fig 2: Scatter plot of the wind power generation vs. generation with natural gas for a month around april 2013.

Fig 2: Scatter plot of the wind power generation vs. generation with natural gas for a month around April 2013. Color indicates power demand at that moment.

What this kind of analysis reveals is that wind power has essentially no correlation with either monthly demand nor with nuclear power production. It does correlate strongly with gas power and less strongly with coal.

To figure out which power source is replacing which we should look at rate of changes in the output. This suggests that further insights may be gained by Fourier transforming into frequency space. Result is demonstrated in Figure 4. Demand shows clear peaks which indicate the familiar regularities. There is a strong peak at zero frequency corresponding to base load demand, there are peaks close to zero frequency corresponding to weekends, and then strong peaks corresponding to a period of about one day.

Nuclear power is strongly peaked at zero frequency demonstrating that it caters for the the base load demand. The peaks in demand have their matches in coal and especially in gas generation. Wind power has a broad featureless spectrum and in order for it to “replace” another power source, this other power source must have appreciable amplitudes at the same frequencies. Only for gas and to a lower extent coal power is this true.

Fig 4: Demand and production in frequency space. (Never mind the units along y-axis)

Fig 3: Demand and production in frequency space. (Never mind the units along y-axis)

In conclusion, UK production and demand data suggest common sense relationships. Wind power acts mainly together with gas while missing EDF reactors were (sadly) mostly replaced by increasing the use of coal.

2.9.2014: Minor changes. The time period after shutdowns, in table with average power levels, was changed to start from thursday 14.8. Previous starting date included the ramp down as well.

Follow me on Twitter


Punainen risti